The Epidemiological Characteristics And Appropriate Diagnostic Cut Points Of Dyslipidemia Among Children And Adolescents From Seven Cities In China

BackgroundDyslipidemia is a major public health issue in China.The blood lipid levels among Chinese population have been risen in recent years.In 2012,the overall prevalence of dyslipidemia in Chinese adults was 40.4%,which was significantly higher than that in 2002.A number of studies have found that dyslipidemia in childhood can cause preclinical atherosclerosis and track into adulthood.Children with dyslipidemia are more likely to suffer from atherosclerotic cardiovascular disease(ASCVD)in adulthood.However,the number of children with dyslipidemia is increasing in China,which indicates that the burden of ASCVD will continue to be increased in the future.Therefore,early prevention and treatment of lipid disorders in children have great significance to the reduction of long-term burden of ASCVD in China.At present,the prevention of dyslipidemia in Chinese children has following challenges:first,the epidemiological data of children's dyslipidemia are limited to some specific regions or ages,and there is still a lack of nationwide investigation data;second,the guidelines recommend dyslipidemia screening in obese children diagnosed with body mass index(BMI).However,BMI cannot distinguish fat mass from lean mass and does not accurately reflect the risk of dyslipidemia that caused by the increased fat mass.There are few studies with the aim of comparing of different obesity indicators' screening effects on dyslipidemia in children;third,the diagnostic cut points of dyslipidemia in Chinese children are the same as the national cholesterol education program and Japanese pediatric criteria.Since the lipid levels vary according to race,gender,and growth,the current cut points may be misdiagnosed when practicing in China.So far,few studies have explored the appropriate cut points for the diagnosis of dyslipidemia in Chinese children.Therefore,using a Chinese multicenter cross-sectional study we aimed to describe the epidemiological characteristics and related factors of dyslipidemia,and explore the diagnostic cut points and high-risk population of dyslipidemia in Chinese children.Moreover,we validated the different sets of cut points in a prospective cohort,which can provide scientific evidence for the prevention of dyslipidemia in China.Objective1.To describe the epidemiological characteristics and analyze related factors of dyslipidemia in Chinese pediatric population;2.To analyze the relationships between body composition and dyslipidemia;3.To compare the effects of different obesity indicators on dyslipidemia screening;4.To explore and validate the appropriate diagnostic cut points of dyslipidemia in Chinese pediatric population.MethodsThe data were obtained from China child and adolescent cardiovascular health(CCACH)study and School-based cardiovascular and bone health(SCVBH)program.CCACH used a multicenter cross-sectional design to conduct a questionnaire survey,body composition assessment,and blood specimen laboratory assay(n=13067)in children aged 6 to 18 years in seven cities in China between 2013 and 2015.Dyslipidemia prevalence was described according to different characteristic groups.Multiple linear regression,unconditional logistic regression,and generalized additive model were used to analyze the associations of body composition,body fat distribution and cardiovascular metabolic risk factors with dyslipidemia.Sensitivity,specificity,positive predictive value,negative predictive value,and area under the receiver operating characteristic curve(AUC)were used to compare the effects of different obesity indicators on dyslipidemia screening.After excluding individuals with extremely obese,severe cardiovascular and endocrine diseases,12875 children were used to develop the dyslipidemia diagnostic cut points using general additive model for location scale and shape(GAMLSS).Generalized Akaike information criterion(GAIC)was used to assessed the goodness of fit,and the final models were verified by residual analysis.Paired ?2 test was applied to examine the agreement of dyslipidemia diagnosed by developed and expert consensus cut points.SCVBH was a prospective cohort study.Stratified cluster sampling method was used to select children aged from 6 to 16 years in Beijing.A total of 14396 children completed the questionnaire survey,physical examination,and blood biochemical assay in November 2017 for the baseline survey,then 10809 children of them have completed the follow-up investigation in November 2019(follow up rate:75%).The capability of baseline dyslipidemia diagnosed by different cut points in predicting persistent dyslipidemia,hypertension,hyperglyciemia,excess body fat,obesity,and hepatic dysfunction that occurred in follow-up period were compared using sensitivity,specificity,positive predictive value,negative predictive value,and AUC.Results1.The prevalence of dyslipidemia in the CCACH population1.1 The average levels of total cholesterol(TC),low-density lipoprotein cholesterol(LDL-C),high-density lipoprotein cholesterol(HDL-C),triglyceride(TG)and non-high-density lipoprotein cholesterol(non-HDL-C)in CCACH population were(3.82�0.73,2.13�0.61,1.41�0.29,0.70[0.50-0.99],and 2.41�0.66)mmol/L,and the prevalence of abnormal lipid levels were 4.2%,3.4%,8.6%,5.0%,and 3.5%,respectively.The prevalence of combined dyslipidemia was 16.7%.1.2 The dyslipidemia was mainly characterized by low HDL-C(boy:9.9%;girl:7.1%)and high TG(boy:5.4%;girl:4.6%),and the prevalence of them were significantly higher in boys and north(P<0.05).Children who attempted to smoke had an increased risk of having low HDL-C(OR[95%CI]:boy,1.65[1.09?2.51];girl,1.71[1.10?2.65]).The risk of having high non-HDL-C was reduced in boys who achieved the ideal medium and high intensity exercising(OR[95%CI]:0.63[0.40?0.99]).1.3 Children with hyperglycemia had higher risk of having high TC,LDL-C,and non-HDL-C(P<0.05)than that with normal level of fasting plasma glucose.Insulin resistance was associated with an increased odds of having high TG(OR[95%CI]:boy,1.98[1.50?2.60];girl,2.25[1.69?2.99]).2.Association of body composition and fat distribution with dyslipidemia2.1 Fat mass index(FMI)was positively correlated with TC(?boy:0.296;?girl:0.145),LDL-C(?boy:0.355;?girl:0.194),TG(?boy:0.289;?girl;0.143)and non-HDL-C(?boy:0.391;?girl:0.214),and negatively correlated with HDL-C(?boy:-0.191;?girl:-0.118),especially in boys(P for interaction<0.05).Lean mass index(LMI)was negatively related with TC(?boy:-0.155;?girl:-0.148),LDL-C(?boy:-0.096;?girl:-0.114)and non-HDL-C(?boy:-0.061;?girl:-0.081).2.2 Regional fat mass was positively correlated with the increased odds of dyslipidemia in boys(P<0.001),especially in android(OR[95%CI]:1.96[1.74?2.22])and trunk fat mass(OR[95%CI]:1.97[1.74-2.23]).There were no significant associations of gynoid and extremity mass index with the risk of having high TC,LDL-C and non-HDL-C in girls(P>0.05).3.Comparison of different obesity indicators in dyslipidemia screeningThe sensitivity,specificity and AUC(95%CI)of FMI in dyslipidemia screening were 47%,79%and 0.627(0.603?0.652)in boys,and 42%,77%and 0.591(0.565?0.617)in girls,respectively.The predictive capability of FMI was significantly higher than that of BMI(PAUC<0.05).4.Validation of different dyslipidemia diagnostic cut points4.1 By linking to Chinese adults dyslipidemia cut points at age 18,the recommended centiles for defineing abnormal TC,LDL-C,TG,non-HDL-C levels were P93-P98 among Chinese children aged beweeten 6 and 17,and P12.4(boy)and P5.0(girl)for abnormal HDL-C,respectively.4.2 Compared with the expert consensus cut points,the developed cut points increased the prevalence of low HDL-C and high TG in boys,and decreased the prevalence of all types of dyslipidemia in girls.There was no significant trend of combined dyslipidemia prevalence diagnosed by the developed cut points with age(boy:?2 for trend=1.209,P=0.233;girl:?2 for trend=-0.748,P=0.457),while the prevalence of combined dyslipidemia diagnosed by expert consensus was increased(?2 for trend=-3.719,P<0.001)in boys and decreased in girls(?2 for trend=3.211,P=0.002)with age.4.3 A total of 701 children among SCVBH baseline population had different classifications when diagnosed using the two sets of cut points.Compared to those with dyslipidemia diagnosed by the expert consensus cut points,the children with dyslipidemia diagnosed by the developed cut points had increased risk of obesity at the follow-up stage(RR[95%CI]:2.16[1.30?3.59]).4.4 The capability of baseline high TG in predicting of hypertension(AUC:0.526 vs 0.519)and excess body fat defined by fat mass percentage(AUC:0.523 vs 0.514)at the follow-up were better using developed cut points(PAUC<0.05)among boys.There was no significant difference between two sets of cut points in the prediction of non-lipid cardiovascular metabolic risk factors in girls(PAUC>0.05).4.5 The AUCs of the prediction in persistent dyslipidemia using developed cut points were significantly higher than that using expert consensus in girls(PAUC<0.05).In addition,the developed cut points had better capability in predicting persistent high TC and non-HDL-C among boys(PAUC<0.05).Conclusions1.The prevalence of dyslipidemia among children aged 6 to 18 years in seven cities was 16.7%.The dyslipidemia was mainly characterized by low HDL-C and high TG which were higher in boys and north.Attempting to smoke behavior is associated with low HDL-C in children.2.The associations of body fat and lean mass with blood lipids are different in children.Body fat is positively correlated with TC,LDL-C and non-HDL-C,especially in boys,while lean mass is negatively correlated with them.3.The android and trunk fat mass are associated with an increased risk of dyslipidemia among children.Gynoid fat mass is not correlated with the risk of dyslipidemia in girls.4.The dyslipidemia screening effect in the population with excess body fat defined by FMI is better than that of obese people defined by BMI.5.This study explored the appropriate diagnostic cut points of dyslipidemia for Chinese children aged 6 to 17 years,which took into account the physiological fluctuations of lipid level and were linked to the adult cut points to establish an indirect correlation with long-term cardiovascular outcome events.The developed cut points could be used in dyslipidemia screening among Chinese pediatric population.6.The capability of developed cut points in the prediction of persistent dyslipidemia is better than the expert consensus.The cut points should be further verified by more large samples.